© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 107349
    Prev
    Next
    INDEX
    Explanations

    references to harm or negative impacts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    harm

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Administrativna
    -0.67
    Jeografia
    -0.61
     Sixt
    -0.52
    uxedo
    -0.52
    estanden
    -0.51
    ⓧ
    -0.51
     PUC
    -0.51
     panty
    -0.50
    Dix
    -0.50
     Dix
    -0.50
    POSITIVE LOGITS
    harm
    0.96
     harm
    0.96
     Harm
    0.91
    Harm
    0.89
    Harmful
    0.88
     harms
    0.84
     Harmful
    0.81
     harmful
    0.81
     harming
    0.79
     daños
    0.74
    Activations Density 0.011%

    No Known Activations